Schrödinger KNIME Extensions

Schrödinger KNIME Extensions New Features

https://www.schrodinger.com/sites/default/files/knime_newfeatures.pdf

Help and support materials

•  Product page http://www.schrodinger.com/KNIME-Extensions •  Schrödinger Nodes

Listed at the end of the New feature presentation https://www.schrodinger.com/sites/default/files/knime_newfeatures.pdf

Node descriptions accessible from Help > Help content > Knime > Node descriptions •  Workflow examples http://www.schrodinger.com/knimeworkflows

Including simple ones in Nodes and advanced ones in Real life group Major improvements listed in the New feature presentation

•  Schrödinger extensions Manual https://www.schrodinger.com/documentation Overview http://www.schrodinger.com/upload/KNIME_Overview.pdf FAQs http://www.schrodinger.com/kb

Help and support materials

Installation instructions and download links Update site, zipped version, free nodes, installer script

http://www.schrodinger.com/kb/1085 •  KNIME.org and KNIME.com

Learning hub http://www.knime.org/learning-hub?src=knimeapp List of KNIME Nodes and Features

http://www.knime.org/files/node-documentation/index.html Example workflows https://www.knime.org/example-workflows

including publicserver_workflows.zip Forum https://tech.knime.org/forum KNIME.com products https://www.knime.org/products/product-matrix

•  Contact us: [email protected]

•  Includes the latest version of KNIME (v3.5) •  New Bioisoster replacement node •  New Quick form nodes supported in KNIME workflows in Maestro•  Antibody building, Sequence converter metanodes •  Reduced the python panels (eg LigPrep node) opening time on Windows and

2D renderer speed up

[KNIME Schrödinger extensions version 4.1]

2018-1 New Features

KNIME Analytics Platform 3.5 – Some New Features

KNIME Big Data Extensions –  The Create Spark Context Node

–  The Spark Missing Value Node

–  The Spark GroupBy Node

KNIME Analytics Platform •  JavaScript Views (Tag Cloud, Data Explorer, Table Editor)

•  New Integrations –  KNIME Deep Learning - Keras Integration

–  Google Sheets Nodes

–  Run R Model in Microsoft SQL Server

–  Improved H2O Integration

•  Extensions that have “graduated” from KNIME Labs (Text Processing, REST Client Nodes, some JavaScript View Nodes, Python Integration, supporting Python 3)

•  New Utility Nodes –  The Math Formula (Multi Column) node

–  The OPTICS clustering nodes

–  The Window Loop Start node

KNIME Server –  OpenAPI definitions of individual workflows

–  More functionality exposed as REST resources

See details on https://tech.knime.org/whats-new-in-knime-35, the video

and the full list of changes in the https://tech.knime.org/changelog-v35


•  In Wrapped metanode, the node usage and layout in the KNIME portal is easier to control

•  With a standalone installation, waiting for KNIME 3.5.2 fix, the verbosity has to be changed in workspace

Bioisoster replacement node

•  Node configuration with the same panel as in Maestro •  Under Library design

AutoQSAR – DeepChem

•  New options available in the nodes using Maestro Python configuration pane –  eg DeepChem method in the AutoQSAR node

Minor Improvements and Fixes

•  Performance improvements –  Reduced the python panels (eg LigPrep node) opening time on Windows

–  2D renderer speed-up

•  Induced fit docking node configuration panel couldn’t be restored between releases

•  ConfGenX node robustness improvement •  KNIME-Maestro Connector node bugs •  New Quick form nodes supported in KNIME workflows in Maestro

–  To take advantage of the Wrapped metanodes

New and Improved Workflow Examples

•  Sequence converter –  New workflow under Protein category

•  Bioisoster replacement –  New workflow under Library design

•  Bioluminate Build antibody –  New workflow under Protein category


•  Bioluminate Build antibody New workflow under Protein category


•  Python script execution (under General – Python) A python script stored in the workflow in run in the Chemistry external tool node. The Python script can also be stored in the Table creator node to be edited easily

2017-4 New Features

•  The Residue scanning, Conformer cluster and AutoQSAR nodes are more robust •  KNIME preferences can be changed and other option passed when run from Maestro [KNIME Schrödinger extensions version 4.0]

KNIME preferences when run from Maestro

•  KNIME preferences can be changed and other option passed when run from Maestro –  SEE STU test

•  https://stu.schrodinger.com/test/28259 •  https://jira.schrodinger.com/browse/KNIME-4734 •  https://jira.schrodinger.com/browse/KNIME-4759


•  RMSD and Phase screening nodes •  Update site •  KNIME in Maestro workflow examples

2017-3 New Features

•  Includes the latest version of KNIME (v3.4) •  New Covalent docking node •  New Prime loop sampling node •  New Get Help node •  Desmond System Builder and Molecular Dynamics nodes have updated

configuration panels [KNIME Schrödinger extensions version 3.9]


•  KNIME Personal Productivity now part of KNIME Analytics Platform Now free Local Metanode templates, call local workflows inside of another KNIME workflow, set up your own version of KNIME’s Workflow Coach, and use the Workflow Diff to show the differences between workflows or different versions of the same workflow. Select 2 workflows in the KNIME explorer window and Compare in the pop-up menu •  Wrapped metanode composite view The default view for wrapped metanodes containing quickforms and/or JavaScript views now shows the same view you would see if the wrapped metanode were opened with the KNIME Web Portal. This opens up many new possibilities for interactive data analysis in KNIME. •  Logistic Regression nodes are more scalable, faster and support regularization •  A new version of the Python integration •  New Date & Time integration •  Integration with the H2O machine learning library •  Audio and speech recognition nodes •  New JavaScript Views •  New Cloud Connectors KNIME Server 4.5: “Deploy to Server” and “Open in WebPortal” menu items KNIME Big Data Extensions: Cloud connectors for common big data file formats, Support for Spark 2.0 See details on https://tech.knime.org/whats-new-in-knime-34, the video https://www.youtube.com/watch?v=cex1xzq5OC8&feature=youtu.be and the full list of changes in the https://tech.knime.org/changelog-v34

Desmond System Builder and Molecular Dynamics

•  Updated the types, tools and the node configuration dialog with the same panel as in Maestro

Covalent docking

•  Node configuration with the same panel as in Maestro •  Under Docking and scoring:

Prime loop sampling

•  The node has a simplify configuration panel but all the GUI and command line options can be controlled through Parameter flow variables. eg a string flow variable RefineLoops--PRIME_TYPE set to EXTENDED See all the options in: https://www.schrodinger.com/sites/default/files/s3/mkt/Documentation/2017-3/docs/Documentation.htm#prime_command_reference/prime_commands_prime_input.htm

Get Help

•  Shows command line options that can be passed through parameter flow variables –  For a node (via the backend flow variable) or tool specified in the node.


•  Mae-to-SD and SD-to_smiles nodes support V3000 format •  Extract properties supports the new Phase type as input •  Ligprep

–  Fixes in the new node, before removing the deprecated old node

•  Chemistry External Tool- Sequence and Alignment properties not populated in Properties tab •  AutoQSAR Build Model

–  Error when executing the node with custom descriptors


•  Desmond System Builder and Molecular Dynamics –  MD – MD simulation: updated workflows using the new nodes

•  Covalent docking –  Docking – Covalent docking: new workflow

•  Prime loop sampling –  Protein – Model building: additional branch using the new node and illustrating the Parameter flow variable control

2017-2 New Features

•  New AutoQSAR node •  Confgen node uses Confgenx fast 3D Conformation generator •  KNIME_batch.py takes .knwf workflow extensions •  Avoid confusing Potential deadlock in SWT display thread warning message in

KNIME 3


AutoQSAR node

•  Node configuration with the same panel as in Maestro

•  The node is in Labs category for now as it may have a few rough edges

New Workflow Example

•  AutoQSAR node –  Cheminformatics – AutoQSAR

2017-1 New Features

•  Includes the latest version of KNIME (v3.3.1) •  LigPrep node has an updated configuration panel •  Workflow list node reports the script nodes and lists the flow variables

controlling the node configuration •  KNIME-Maestro Connector node has a new Phase Hypothesis mode •  Chemistry External Tool node supports both Hypothesis formats [KNIME Schrödinger extensions version 3.7]

2016-4 New Features

•  Phase screening node uses the new configuration panel and new algorithm •  Any keyword can be passed to the Jaguar nodes •  Extract properties node supports Sequence and Alignment inputs •  KNIME 3.x zipped update site is available in the installation [KNIME Schrödinger extensions version 3.6]

New Phase Screening Node

•  Node configuration with the same panel as in Maestro •  The new node uses the new algorithm to screen compounds from Phase

databases or files. Standard types (mae, sdf) are supported as input and output

•  Phase Hypothesis Reader and Writer nodes support new (.phypo and .mae) and old (.xyz) Hypothesis formats

Jaguar and Extract Properties Nodes

•  Any keyword can be passed as Parameter flow variable to the Jaguar nodes –  eg INFO Jaguar Minimization Detected a JaguarMinimization flow variable: JaguarMini--iguess=30

•  Extract properties node supports Sequence and Alignment inputs

KNIME 3.x Zipped Update Site

•  KNIME 3.x compatible version –  The zipped update site available is in the installation under $SCHRODINGER/knime-v*/data/

(along with the KNIME 2.12 version) –  It is also available from our Update site and the zipped update site repository: https://support.schrodinger.com/releases/knime/extensions/ and https://support.schrodinger.com/releases/knime/zippedUpdateSite/

Improved Workflow Examples

•  Phase screening –  Pharmacophore – Phase screening: use the new node

Same node for Phase database and file screening •  Extract properties supports Sequence and Alignment inputs

–  Protein – Model building: chose alignments in a list with corresponding properties

2016-3 New Features

•  KNIME 3.x is officially supported (separate zipped and update site version compatible)

•  Induced fit docking has an updated configuration panel •  In Maestro 11:

–  Establish structure exchange with KNIME moved to Tasks > Workspace and Project table operations > KNIME –  The workflow menu has been merged with the Script menu


KNIME 3.x and Induced fit docking node

•  KNIME 3.x is officially supported –  Specific versions available on the update site https://support.schrodinger.com/releases/knime/extensions/

and https://support.schrodinger.com/releases/knime/zippedUpdateSite/

–  The nodes supporting mol2 and sdf input and output adapter cells •  Induced fit docking node

–  Updated node configuration with the same panel as in Maestro

KNIME menus in Maestro 11

•  Establish structure exchange with KNIME has been moved to Tasks > Workspace and Project table operations > KNIME

•  The workflow menu has been merged with the Script menu where KNIME workflows can now be imported and run in Maestro

2016-2 New Features

•  Minor improvements in the Shape Screening node •  The Python Script nodes pass unsupported types to the output table •  Fixed output column row matching when some rows are missing •  Minor improvements in the Extract Atom/bond Properties node [KNIME Schrödinger extensions version 3.4]

New features

•  Shape screening –  Minor improvements and the new version of the node has been moved to the Pharmacophore modeling category

•  Python Script nodes –  Pass unsupported types to the output table

•  Fixed output column row matching when some rows

are missing –  eg failed structures, “duplicate row ID” error message

•  Minor improvements in the Extract Atom/bond Properties node –  Handle missing input cells, Populate the Properties tab in output tables

2016-1 New Features

•  Still includes KNIME 2.12 but a KNIME 3.1 compatible version is available for testing •  Shape screening has an updated configuration panel and URLs are supported •  The parameter flow variables can be used in Strain Rescore node and more Canvas

nodes •  Short property names and Include input available in the Extract Atom/bond Properties

node [KNIME Schrödinger extensions version 3.3]

KNIME 3.1- some New Features

•  GUI and general Structure –  New Look and Feel, New Eclipse and Java (Java 8, Eclipse 4, and BIRT 4) –  WrappedNodes

–  Node Repository - Fuzzy Search

•  Analytics and ETL –  Simple Regression Tree, Random Forest , Active Learning , ARIMA, k-Means, Prediction Fusion, Rank (New Nodes)

–  More In-Database processing nodes

•  Streaming: Executor, Text Processing Nodes •  KNIME Server: REST interface, Cloud Server (license

required)

•  KNIME Personal Productivity : WorkflowDiff (license

required) See details on https://tech.knime.org/whats-new-in-knime-31 and the full list of changes in the changelog

KNIME 3.1 and Shape Screening

•  A version compatible with KNIME 3.1 is available for testing: https://support.schrodinger.com/releases/knime/zippedUpdateSite/SchrodingerKNIMEUpdateSite_KNIME3.x.zip

–  sdf and mol2 input columns aren’t supported yet: a Molecule to Maestro converter node has to be used first.

•  New Shape screening node –  Updated node configuration with the same panel as in Maestro –  Available in the Labs category and old version still available but deprecated soon –  URLs supported and editable for an external file to screen (not for a Phase database)

eg knime://LOCAL/… and file:///path/file


•  The parameter flow variables can be used in Strain Rescore node and more Canvas nodes: Generate Pairwise Matrix, Generate Pairwise Matrix (2 inputs), Similarity Matrix

•  Short property names and Include input available in the Extract Atom/bond Properties node

•  Reader nodes support spaces in the path •  Run Maestro Command supports sdf input •  Ligand size and diameter fields no longer ignored

in the Glide Grid Generation node


•  Shape screening –  Pharmacophore – Shape screening: new node

•  Parameter flow variable support –  Labs – Parameter flow variables: Force field control for the Glide ligand docking, Strain rescore, Canvas similarity search (to be run on a remote fingerprint file), Generate pairwise matrix, Similarity matrix nodes.

For the nodes using a Python configuration panel, flow variables can't be used to control the options in the panel. However Parameter flow variables can be used to control the equivalent command line options.

This is applicable for the following nodes: Shape screening (labs), IFD (labs), Residue scanning, Conformer cluster.

•  Extract atom/bond properties –  Nodes – Node tools: new feature usage

2015-4 New Features

•  Includes the latest version of KNIME (v2.12.0) •  Faster and more robust Extract mae atom/bond properties and Extract Maestro

properties nodes •  A configured reader node is added when dragging-and-dropping a Grid, Phase

hypothesis file into the workspace •  URLs supported and editable in the Molecule Reader node. The output column name

is editable •  The Desmond molecular dynamic and Chemistry external tool nodes accept a

trajectory as input •  Size property in the Group MAE node [KNIME Schrödinger extensions version 3.2]

2015-3 New Features

•  Compatible with the latest version of KNIME (v2.12.0) •  Better control on calculation distribution

–  Automatically use the maximum number of processors available on the machine –  Easily change the machine some nodes in a workflow are run on

•  New Rotate all and Protein Structure Alignment node improvement •  A configured Molecule reader node is added when dragging-and-dropping a Fasta or a

Smiles file into the workspace •  Zipped update site available in the installation •  OPLS3 available in the configuration panel force field list •  Use the same Maestro session to run commands at several stages of the workflow


KNIME 2.12 – some New Features

It was released too late to be included. It will be in 2015-4.

•  Log messages in workflow associated with workflow and nodes

•  Analytics

–  Rule Handling

–  Statistics measure as aggregation methods in GroupBy node

–  More Statistics Nodes

•  Tool Integration

–  Javascript Integration (new interactive visualization nodes)

–  Extended Python Integration (incl. Python edit variable)

•  Utility Nodes

–  CASE Switch (eg taking a flow variable as input)

–  GUI Improvements

•  Drag&Dropping nodes from the Node Repository over a node or over a connection

•  Hotkeys are now available in the Context menu

•  KNIME Server & Automation

–  REST interface

–  Call Local / Remote Workflows from within another workflow

–  KNIME Local Automation (New product)

See details https://tech.knime.org/whats-new-in-knime-212 where you can also find the full list of changes.


Better Control on Calculation Distribution – Preferences

•  Primary and secondary hosts –  Define PRIMARY_HOST and SECONDARY_HOST placeholders to be used in the Jobcontrol tabs –  Set the maximum number of processors to be used when the number of processors is set to 0 in the node Jobcontrol tab. If set to 0 in the Preferences all the processors on this machine will be used.

•  Default host for new nodes –  The list includes PRIMARY_HOST and SECONDARY_HOST placeholders

Jobcontrol Panel settings

•  #CPUs –  If set to 0 the node uses the maximum number of processors on the host or set in the Preferences

New default for the new nodes. Use the Primary host as Default host for new nodes to mimic the old behaviour.

•  Hostname list –  Includes the PRIMARY_HOST and SECONDARY_HOST placeholders set in the Preferences

•  Separate jobs into subjobs: –  If turned off it uses the same number of subjobs as the number of processors

Start-up script options

•  Corresponding start-up script options –  –primaryHost (default is localhost), –primaryCPU (default is 0), –secHost and –secCPU –  The Preferences stored in the workspace are overridden by these options

Better Control on Calculation Distribution – Use cases

•  Workflow to be used on different machines or shared with colleagues –  Nodes set to use localhost and 0 as number of processors

The host and the maximum number of CPUs specified in the KNIME workspace the workflow has been imported in are used: e.g. my cluster – 12 processors or localhost – 4 processors

•  Test a workflow locally before running it on a cluster –  To test the workflow set computationally expensive calculation nodes to run on the Primary host and 0 as #CPUs in the Jobcontrol tab.

In the Preferences set the Primary host to localhost and 2 processors for instance. –  To run the workflow on a large data set, in the Preferences, change the Primary host to a cluster and 10 processors

Reader Nodes from a Dragged Smiles or Fasta File

•  A Sequence node is added and automatically configured when dragging and dropping a fasta from a file explorer or the KNIME explorer view into the workspace. A Smiles reader node is added for Smiles files.

•  This has already been possible for Maestro files associated with the Molecular reader •  Text (.csv and .txt), Mol2 and sdf files are associated with the KNIME Analytics platform

reader nodes

Rotate All and Protein Structure Alignment Nodes

•  They both align protein structures as rigid bodies, making use of secondary structure information. The Rotate all node initially aligns the first mobile protein and apply the transformation to the remaining mobile structures.

•  The alignment score and RMSD are reported and can be used for filtering

OPLS3 in the Force Field List

•  The force field can be set to OPLS3 in the configuration panel force field drop down list

•  Reminder: the configuration panel options can be controlled by simple flow variables and don’t require Parameter flow variables


•  Use the same Maestro session to run commands at several stages of the workflow –  Using pythonrunbuiltin KNIME_maestro_connector.send command

•  Zipped update site available in the installation:

–  $SCHRODINGER/knime-v*/data/SchrodingerKNIMEUpdateSite_3.1.7.201506160750.zip –  Easy access to add the extensions to or update a stand-alone installation

KNIME_install.py could also help installing a set of extensions (not only Schrodinger ones) –  Also available on:

https://support.schrodinger.com/releases/knime/zippedUpdateSite/

•  The Shape screening node runs the new fast algorithm. The old method can still be used with a the Parameter flow variable ShapeScreen-classic

•  Better logging in Protein Preparation Wizard •  Molecule-To-MAE node reports the exit status of the

converter •  Preprocessing command script section added to

Run PyMOL


•  Better control on calculation distribution –  All workflow examples are set to run on the Primary host and the number of CPUs specified in the Preferences. So these can easily changed to run on your machine of choice and number of CPUs. –  Docking - Docking: illustrates all the new functionality

•  Rotate all and Protein Structure Alignment –  Protein- alignment workflow example

•  Smiles and Fasta readers –  Used in Protein - Model building and Cheminformatics - Clustering workflows for instances.

•  OPLS3 –  Lab - Parameter flow variable use-cases: the force field can be set to OPLS3 using a simple flow variable

•  Use again the same Maestro session to run commands –  General - Run Maestro command use cases: so Maestro window can be put in the background

•  Phase Shape –  Lab - parameter flow variable use-cases: to use the old algorithm

2015-2 New Features

•  Includes the latest version of KNIME (v2.11.3) •  New Multiple Sequence Viewer node •  Molecule reader node URL editable to use “knime:” and “file:” •  The parameter flow variables can be used in LigPrep, Molecule reader and

Molecule-to-MAE nodes •  Easy way to list the command line options to be used as Parameter flow variables •  Free nodes don’t require a Schrödinger installation when possible •  Residue scanning and Conformer cluster configuration panel start-up faster •  Run Maestro command and Jaguar minimization nodes are more robust [KNIME Schrödinger extensions version 3.0]

Multiple Sequence Viewer

•  Sequence and alignment visualization and manipulation

•  Command section –  Allows for automation of some tasks –  See the list of the commands in the node description

including send_to_knime for headless use of the MSV

Molecule Reader Editable URL

Molecule reader node URL is editable so as to point to:

•  A file embedded in the workflow

–  knime://knime.workflow/<directory in the workspace workflow folder>/<filename>

–  Usage: embed input files

–  Get the path of the workflow folder from the pop-up menu in the KNIME explorer > Copy Location

•  A file in the workspace

–  knime://LOCAL/<workflow group(s)>/<filename>

–  Similar to dragging and dropping a file stored in the workspace

•  A file on disk

–  file:/<path on disk>/<filename>

–  Similar to dragging and dropping a file from a file explorer in the workflow

•  First configuration with a (possibly mock-up) file

Parameter Flow Variables and Option List

•  The parameter flow variables can be also used in LigPrep, Molecule reader and Molecule-to-MAE nodes

•  Easy way to list the command line options that can be used as Parameter flow variables: –  Use the backend flow variable available in these nodes


•  Residue scanning and Conformer cluster configuration panel start-up faster

•  Run Maestro command and Jaguar minimization nodes are more robust

•  Plugin version number changed to: mmshare version – date See Help, About KNIME, Schrödinger icon

Some Free Nodes don’t Require a Schrödinger Installation

They don’t require any Schrödinger installation when possible and otherwise only a Maestro installation

•  Reader, Writer and Converter nodes •  Visualization: Run PyMOL, Run Maestro, KNIME-Maestro connector,

Multiple sequence viewer, Run Spreadsheet viewer, Table viewer •  Scripting: Chemistry external tool node, Run Maestro command •  Tools: Workflow List, Setup Diagnosis nodes •  Utility script: KNIME_install.py See details in:

https://www.schrodinger.com/knimeworkflows/KNIME_Free_Nodes.pdf

Requirement and Setup

•  For the Workflow List, Run Spreadsheet viewer, Table viewer, Chemistry external tool nodes: –  Install Schrödinger free nodes:

https://www.schrodinger.com/SchrodingerKNIMEFreeNodes

http://tech.knime.org/community - update site: http://tech.knime.org/update/community-contributions/2.11/

On Windows the Chemistry external tool and Workflow list nodes require specific set-up (see KB#1925)

•  For the Run PyMOL node: –  In Workspace Preferences, point to any PyMOL installation

•  For all the other free nodes and the utility scripts: –  In Workspace Preferences, point to a Maestro installation

Maestro is free of charge for all academic users and commercial users who purchased one of the Schrödinger tools

Installer and license from https://www.schrodinger.com/freemaestro/ (also includes the zipped update site)

–  The install script is available from https://www.schrodinger.com/KNIME_install/KNIME_install.zip


•  Multiple Sequence Viewer –  Nodes – MSV: simple example –  Protein – Model building and Real world examples – Protein model building: new node use cases

•  Parameter flow variables –  Lab – Parameter flow variables: illustrates more node use cases. Based on the backend flow variable the command line options that can be used are listed.

•  Molecule reader node accepts “knime:” and “file:” URLs –  Nodes – Readers and Writers: simple example –  Pharmacophore – Shape screening and many other workflow examples: point to input files stored in the workspace –  Protein – Model building: same URL format used to use a script embedded in the workflow


•  KNIME UGM workflow –  Real world example – Cocrystallized ligand redocking: improved ligand detection process

Derived from the work published by Paulette Greenidge from Novartis in: J. Chem. Inf. Model. 2014, 54(10), 2697–717

•  Free nodes –  Nodes – Free nodes: specific examples using only free nodes.

•  Simple workflow examples –  Nodes – Reader and writer nodes: including reading a list of PDB files in a folder –  Nodes – Run Maestro and Run Maestro commands –  Nodes – Workbench: list of the most commonly used nodes

2015-1 New Features

•  Includes the latest version of KNIME (v2.11.0) •  KNIME-Maestro Connector improvements to replace the Run Maestro 1:1 metanode •  New Conformer cluster node

•  Run the Workflow list node on a selected workflow group •  The simplified batch execution script KNIME_batch.py must now be run without

-FROM maestro •  The Schrödinger extensions in a stand-alone KNIME installation only requires setting

$SCHRODINGER environment variable •  Simplified KNIME_install.py invocation to install all community nodes •  A specific executable file can now be chosen in the Run PyMOL node installation field



•  Database and Big Data Extension

Database GroupBy improvement, Drop Table, HP Vertica Connector, Impala Connector / Loader, HDFS Connector / File Permissions new nodes

•  Tool Integration

–  JSON Processing

–  New Python Integration

•  Data Mining

Modular PMML, DBSCAN density based clustering algorithm, kNN now supports more distance functions, Target Shuffling

•  IO – Writer Nodes Improvements

•  Other

–  Quick Node Insertion with Ctrl-Space (New GUI feature)

–  Table Validator and Column Auto-Type Cast new nodes

See details in http://tech.knime.org/whats-new-in-knime-211 and full list of changes in changelog file


•  New Python Integration –  based on CPython –  Schrödinger Python installation can be used after installing Protobuf (for the communication between CPython and KNIME) and optionally Jedi (for the auto-completion)

•  Column Auto-Type Cast tries to guess the most fitting type

for a specified column. It is useful after a Transpose and before transforming a data cell into a flow variable.

•  Table Validator checks the input table format

against missing values, out of domain values…

Path, Environment, and Update Simplifications

•  Simplified batch execution script KNIME_batch.py must now be run without -FROM maestro:

eg $SCHRODINGER/run KNIME_batch.py -stdout INFO "D vs 1.zip" -schrodingerTempDir /tmp/KNIME -run -isdf ligands.sdf -igrd fxa_grids_recep -osdf /tmp/out.sdf

•  A stand-alone KNIME installation now only requires setting $SCHRODINGER See KB#1696 for other ways to run Schrodinger nodes in a stand alone installation

•  Simplified KNIME_install.py script invocation –  e.g. 1 command line to add all the trusted community nodes (http://tech.knime.org/community) to the embedded KNIME installation: "$SCHRODINGER/run" -FROM knime KNIME_install.py -features_list "$SCHRODINGER/knime-v*/data/full_free_features.lst" "$SCHRODINGER/knime-v*/bin/*/knime/“ –  The KNIME version embedded in $SCHRODINGER/knime-<version>/bin/<os>/knime

•  Is identical to the download from KNIME web site •  Only some basic extra features are added to keep the size under control

but others can easily be added from the GUI or using KNIME_install.py

KNIME-Maestro Connector Improvements

•  KNIME-Maestro Connector improvement to replace the Run Maestro 1:1 metanode

•  Make some selections/structure changes interactively

•  New modes: –  All entries, Selected entries –  CombiGlide core definition –  Atom selection (as List, Names, ASL, SMARTS) –  Residue selection (as List, ASL)

•  Command script section •  Iterate over entry structures •  Dockable panel

Maestro - KNIME structure exchange (since 2012) KNIME

Projecttable

Maestro Server http://www.schrodinger.com/knimeworkflows/Validatedockingparameters

New Conformer Cluster Node

•  The new Tools > Conformer Cluster node is configured with the same panel as in Maestro

•  Interactive or automatic selection of the number of clusters

–  Visually or based on the Kelley penalty

•  Same look and feel as the other KNIME nodes –  Cancel, Reset, Job control distribution ...

–  Parameter flow variables, batch execution

Minor Improvements

•  Run the Workflow list node on a selected workflow group Faster execution and easier post-processing

•  A specific executable file can be chosen in the Run PyMOL node installation

field


•  KNIME-Maestro connector node replacing the Run Maestro 1:1 metanode in: –  Labs – Run Maestro 1:1 metanode use-case example moved to General – KNIME-Maestro Connector (residue selection, create more entries)

–  General – Python script, Molecular Mechanics – Conformational search and Molecular Dynamics – Simulation (atom selection for distance measurement)

–  Quantum Mechanics – Conformational search (dihedral angle selection)

–  Library design – Library enumeration (core attachment point selection)

–  See also Docking – Validate parameters – KNIME-Maestro connectors (initial mode: existing Maestro session)


•  Conformer cluster –  Nodes – Conformer cluster (simple usage examples) –  Molecular Mechanics – Conformational search (new node application)

•  Real world applications – Co-crystallized ligand docking, derived from a collaboration work with Paulette Greenidge:

“Improving Docking Results via Re-ranking of Ensembles of Ligand Poses in Multiple X-ray Protein Conformations with MM-GBSA PA Greenidge”, C Kramer, JC Mozziconacci, W Sherman. Journal of chemical information and modeling. 2014, 54 (10), 2697-2717


•  Real world applications – Protein model building

Including a Multiple Sequence viewer metanode and homology model building on a specific template

•  Nodes – Epik

Simple usage examples

•  General – Installation update/creation

Add the trusted community nodes in one command. Can even be used on the running installation

•  General – Batch execution

New workflow illustrating KNIME_batch.py options and execution

•  Pharmacophore – Shape Screening and Docking – Virtual screening and General – Run PyMOL and Workflow list

Updated

2014-4 New Features

•  Includes the latest version of KNIME (v2.10.1) •  New Residue scanning/affinity maturation node that uses the same

configuration panel as Maestro •  Run PyMOL output files can be used by downstream nodes •  Pop-ups no longer block Run Maestro and Run Maestro command nodes •  Improved loading of Schrödinger extensions during KNIME start-up [KNIME Schrödinger extensions version 2.8]

New Residue Scanning/Affinity maturation Node

•  Node configuration with the same panel as in Maestro –  Includes direct access to the online help

•  Same look and feel as the other KNIME nodes –  Cancel, Reset, Job control distribution ... –  Batch execution

New Residue Scanning/Affinity maturation Node

•  Run on several structures –  The mutations are selected on the first input structure –  Used for the other structures (if the residues exist)

•  Flow variables –  Configuration panel tab but not for the Settings panel fields

•  Use Parameter flow variables instead

•  Parameter flow variable –  To control the Settings parameters

•  eg ResidueScanning-refine_mut set to python_minimize

–  Access to command line only options/values •  eg ResidueScanning-loop_options set to RES_SPHERE/5.0,MAX_CA_MOVEMENT/4.0,PROTOCOL/LOOP_BLD

–  All the command line options can be listed using the “backend” flow variable

Reporting Nodes Improvements

•  Pop-ups no longer block Run Maestro and Run Maestro command nodes and preference changes are now local –  No change in these nodes will affect the other Maestro sessions

•  Run PyMOL output files (log, structures, images...) can be used by downstream nodes and temporary files are cleaned up

Using %<flow_variable_name>% in the command script or %PyMOL_basename% added to the Variable Outport

Main Fixes and Minor Improvements

•  Improved loading of the Schrödinger ‘s extensions during KNIME start-up No more delay when the Licence server isn’t responding

•  Better control of Parameter flow variables in advanced scenarios Use the “_rm_” keyword to remove an option with a value, while “_off_” can still be used if

the argument doesn’t have a value

•  Most of the nodes use scripts from the distribution –  eg RMSD, Desmond trajectory extract frames –  No difference with the calculations run from Maestro –  Take advantage of the latest improvements and fixes

•  LigPrep no longer fails to write output in SDF •  Some nodes were fixed to run with UNC paths


•  Residue scanning – improvements and use the new node •  Run PyMOL – load rendered output image and structure files •  SiteMap binding site characterization –

generate Ligand Interaction Diagrams

Residue Scanning/Affinity Maturation

Residue Scanning/Affinity Maturation

2014-3 New Features

•  Compatible with the latest version of KNIME (v2.10.0) •  A configured Molecule reader node is added when dragging-and-dropping a

Maestro file into the workspace •  Improved Simplified batch execution and troubleshooting •  Installation script improvements to install Community Contributions and third-

party extensions •  The 2D renderer is more stable on Linux •  Fixed a bug preventing several Glide grid generation nodes to be run in parallel •  Added an output port to the Setup Diagnosis node

[KNIME Schrödinger extensions 2.7]


•  The extensions are compatible with the latest version of KNIME (v2.10.0)

•  It was released too late to be included in Schrodinger’s installation (still KNIME 2.9.2)

•  Find the full list on http://tech.knime.org/whats-new-in-knime-210

•  New Nodes and New Features –  GUI and Infrastructure

•  New Intro Page

•  Auto Save

–  Database •  Database Connector Nodes

•  Database Table Selector

•  SQL Extract / Inject and Database SQL Executor

•  Database Sorter / GroupBy / Joiner

–  Social Media •  Google and Twitter APIs

–  Data Manipulation •  7 new Distance Nodes

•  Moving Aggregation

•  Date/Time Shift, Time Series Missing Values

–  Data Mining / Statistics •  New View in Statistics Node

•  Time Series Analysis

–  PMML

–  Data Generation –  Textprocessing

–  Image Import •  Read Images, String to SVG

•  Improvements –  Data Manipulation

•  Column Rename

•  Normalizer , One2Many, and Many2One nodes

•  Cross Joiner

•  GroupBy, Pivoting, Rule Engine, and Column Aggregator

•  AutoBinner

–  Data Mining •  ROC Curve

–  Open Street Maps Integration

KNIME 2.10 – Auto-save

•  Auto-save If there is an unexpected break, your workflow will

be recovered in the next KNIME session from the latest auto-save file. Disabled by default.

•  Java update consequence

–  On Windows, a re-installation of KNIME is required, due to the underlying Java update. –  On Mac or Linux, you can simply update your KNIME installation via the File menu.

Molecule reader node from a dropped Maestro file

•  A Molecule reader is added and automatically configured when dragging and dropping a Maestro file from a file explorer or the KNIME explorer view into the workspace.

•  For the latter the files are stored in the workspace and the corresponding URL filename is eg: knime://LOCAL/workflow_group/file.mae

These files are exported with the workflow or workflow group if they are located in the directory. They can be organized and inspected directly from the KNIME explorer view.

See the Pharmacophore > Shape screening first workflow example.

Simplified Batch Execution Improvements

•  Standardized the reader and writer node tag: %file% The old tags are still supported

•  Added support for several structure, Schrodinger types and Excel reader and writer nodes

•  KNIME_batch.py -printcmd node list now includes: –  the node annotation for Input and Output nodes –  the node label for Quick Form nodes

-key <Name:$name, Label:$label> [Default:title] See the Pharmacophore > Shape screening and Docking > Virtual screening first workflow examples. $SCHRODINGER/run -FROM maestro KNIME_batch.py -stdout INFO "D vs 1.zip" -run \ -isdf /tmp/knime_workflow_example_data/ligands/fxa_ligands.sdf \ -igrd /tmp/knime_workflow_example_data/Glide_grids/fxa_1f0r_AB_grids_recep \ -osdf /tmp/DVS.sdf

Simplified Batch Execution – New Supported Nodes

•  %file% Molecule Reader/Writer Now the argument base name is -ostruct

•  %file% SDF Reader/Writer •  %file% PDB, Mol2, Smiles Readers •  %file% Glide Grid Reader/Writer •  %file% Phase Hypothesis Reader/Writer •  %file% Sequence, Alignment Reader/Writer •  %file% Canvas Fingerprint Reader/Writer •  %file% CSV Reader/Writer •  %file% XLS Reader/Writer See details with –printallowed

Keywords for workflows to be run in Maestro: •  %GUIsetting% Double, Integer, String Input QF •  %GUIsettingRADIO% String Radio Buttons QF •  %GUIsettingCOMBO% String Radio Buttons

QF •  %GUIsetting% File Upload Quick form node

•  %selected entries% Molecule Reader •  %included entries% Molecule Reader •  %new group% Molecule Writer •  %maestro message% CSV Writer •  %maestro commands% CSV Writer

Start-up/batch Script Improvements

•  Progress information in the terminal for easier debugging in batch: –  stdout <min_log_level> Sets the minimum log level for the stdout ['DEBUG', 'INFO', 'WARN', 'ERROR', 'FATAL'] –  stderr <min_log_level> Sets the minimum log level for the stderr –  suppressErrors Suppress the error window pop-up

On Windows it is in C:\Users\...\AppData\Local\Schrodinger\appcrash\knime.log for now

•  KNIME and Schrodinger options listed separately in the help message Schrodinger options support relative path and KNIME's ones require absolute paths

See the Pharmacophore > Shape screening first workflow example: $SCHRODINGER/run KNIME_batch.py "P Shape 1" -run -stdout INFO –

suppressErrors…

Installation Script Improvements

•  -features_list - any KNIME or third party features can now be installed in one go by specifying the complete list in the file

•  -updatesites option - specifying a version number (like 2.9) will add both KNIME and Trusted Community Contributions web update sites

•  -updatesites option supports comma separated list of update sites (can be both local zipped update site and web urls) So -knime_updatesite was renamed into -updatesites and -knime_features_list into -features_list. The old option names are still supported.

•  eg "$SCHRODINGER/run" -FROM knime KNIME_install.py -updatesite 2.9,http://tech.knime.org/update/community-contributions/2.7/ -features_list /path/to/feature.lst /path/to/myKNIMEinstallation/

With feature.lst file: org.rdkit.knime.feature.feature.group org.knime.features.quickform.feature.group org.erlwood.knime.feature.feature.group (Lilly’s

nodes) jp.co.infocom.cheminfo.marvin.feature.feature.group com.vernalis.knime.feature.feature.group

See General > Installation workflow example and more details about the script in 2014-2 new feature slides.

Installation Script Improvements – use case

" $SCHRODINGER/run" -FROM knime KNIME_install.py -verbose -updatesites 2.10,http://tech.knime.org/update/community-contributions/2.7/ -features_list <see screenshot> -install_basechem 2.10 -schrod_updatesite /path/SchrodingerKNIMEUpdateSite_27006.zip <KNIME installation>

Setup Diagnosis Output

•  Output port to report the results in an output table for post processing

•  Output text file created (easily sent by email) The file name and location is reported in the console and in the last column of the Output port.

•  Reports whether KNIME is started through the startup script or as stand-alone

Main Fixes

•  Fixed a bug preventing several Glide grid generation nodes to be run in parallel

•  The 2D renderer is more stable on Linux now it is run with a new mechanism And it can be used in a stand-alone installation and no environment variable is required

•  No more environment variable required to use the Chemistry external tool node in a stand-alone installation on Windows

•  Zipped update site included in the Windows installers as well (as in Linux)

2014-2 New Features

•  Includes the latest version of KNIME (v2.9.2) •  A tool to automatically install KNIME features and Schrodinger extensions in an

existing installation •  The Molecular Container Type is supported in the Molecule to Mae node •  A default renderer can be selected for the Maestro columns in the Preferences •  New Prime Energy node •  Better calculation distribution in the SiteMap node •  A toggle button to operate on all the properties in the Extract Mae properties and Set

Maestro properties nodes •  Additional formats that Maestro can import are supported in the Molecule reader node

[KNIME Schrödinger extensions 2.6]

Installation Script

•  A command line tool to automatically install KNIME features and Schrodinger extensions in an existing KNIME installation

•  Available under https://www.schrodinger.com/KNIME_install/KNIME_install.zip and $SCHRODINGER/knime-v*/python

•  KNIME_install.py -schrod_updatesite (-install_basechem) -updatesites (-features_list) •  eg to update the Schrodinger extensions in an existing Desktop KNIME installation:

"$SCHRODINGER/run" -FROM knime KNIME_install.py -schrod_updatesite /path/to/SchrodingerKNIMEUpdateSite_26007.zip /path/to/myKNIMEinstallation/

•  eg to install the Schrodinger and a couple of other extensions in a fresh Desktop KNIME installation: "$SCHRODINGER/run" -FROM knime KNIME_install.py -install_basechem 2.9 -schrod_updatesite /path/to/SchrodingerKNIMEUpdateSite_26007.zip -updatesites 2.9 -features_list /path/to/feature.lst /path/to/myKNIMEinstallation/ With feature.lst file containing for instance: org.knime.features.quickform.feature.group com.knime.features.reporting.designer.feature.group org.knime.features.ext.poi.feature.group (see the feature names in "Install New Software..." > "Install details“

2D Renders and Molecular Container Type

•  A default renderer can be selected for the Maestro column type in the Preferences •  The Molecule-to-MAE node supports the Molecular Container type

–  This is an Adapter cells that contains several representations of the same entity

(eg SDF, CDK, RDKit...). So the conversion

is performed only one time –  It could be extended pretty easily

to other nodes if needed

KNIME start-up script – accepts relative paths

•  $SCHRODINGER/knime : defines the required environment for the Schrodinger extensions in KNIME These can also be set in the Preferences

•  Usual KNIME options are passed and some extra or simplified options; See the complete list using –help; They only affect the current session (eg computer specific settings)

•  eg $SCHRODINGER/knime -data Project/myWorkspace -maxHeap 1024m -maxThreads 3 -schrodingerTempDir C:\tmp\KNIME -knimeTempDir C:\tmp -deleteTempFiles true -ooCmd Excel installation path/Open Office (oocalc) (-batch)

Extract/Set MAE Properties – on all the properties

•  Extract Mae properties – all of them –  Useful when the generated properties can’t be expected (eg LigPrep chirality information)

–  see also Extract properties: none, primary, all (eg in Glide ligand docking node)

•  Set Mae properties – all

Other Nodes

•  SiteMap – better calculation distribution See in the SiteMap.in : PROT sitemap_<>_in_1.mae,sitemap_<>_in_2.mae

•  Prime energy – new node •  Split by structure – support the parameter flow variables

eg SplitStruct-merge_ligands_with_chain = “_on_”

Robustness •  Setup Diagnosis – ensure scratch directories are writable, no schrodinger.host file interference

Main Fixes

•  MacroModel nodes etc – expose the force field selector again –  OPLS 2.1 isn’t exposed though and only accessible using the parameter flow variable.

•  Chemistry external tool- Add support to mol type •  Molecule writer- column containing output file name- add .mae if missing •  ConfGen- generated only one conformation per structure

Installation

•  The Download page includes the workflow examples by default:

http://www.schrodinger.com/downloadcenter

•  Update site links: https://support.schrodinger.com/releases/knime/extensions/

https://support.schrodinger.com/releases/knime/zippedUpdateSite/

http://www.schrodinger.com/SchrodingerKNIMEFreeNodes/

(see the list in http://www.schrodinger.com/kb/1710)

2014-1 New Features

•  Includes the latest version of KNIME (v2.9) •  Can use any command line option not exposed in the Glide ligand docking node

configuration panel •  Can use a specific template in the Prime Build Homology Modeling node •  Can operate more easily on the structures passed to the Run PyMOL node

KNIME 2.9 – Some New Features

http://tech.knime.org/whats-new-in-knime-29

•  Column Selection Framework Extended and Improve New Selection Criteria in Column Filter Node and similar nodes

The Column Filter configuration window now allows to select the input data columns: manually, by type, by name based on a RegEx / Wildcard expression

•  Rule Based Framework Extended and Improved New Rule Engine Node, Rule Engine Variable, Rule Based Row Filter/Splitter

•  Workflow Monitoring,, and Administration –  Send Email (new node)

The Send Email node sends an email to a given email address. It can be used to send updates about the workflow execution status.

–  Save Workflow (new node) Performance –  Save As ... <under a different name> (New File Menu Option) –  Easier Extension Installation

All existing KNIME update sites are now available in the Preferences menu item (File -> Preferences -> Install/Update -> Available Software Sites).

–  Context Extractor Node (new node) The node Extract Context Properties makes some workflow context related properties available at the variable output port, including

current user name, workflow name, and working directory.

KNIME 2.9 – Some New Features

•  Data Generation –  Counter Generation (new node)

The Counter Generation node adds a counter column to the input data table. The counter starting point (Min. Value) and the step size (Scale Unit) can be defined in the configuration window.

•  Better Model Training and Evaluation –  Unified Format for Predictor Nodes, Extended Linear Regression Output, Numeric Scorer

•  Other New Features and Nodes –  Cross Joiner (new node)

Performs a cross join of two tables. Each row of the first table is joined with each row of the second table.

–  Pair Extractor (new node) "Unpivot" a distance matrix by extracting all pairwise distances from a distance matrix column.

–  Flow Variables and Missing Values

Glide Ligand Docking

•  Can use any Glide Simple Input file option not exposed in the configuration panels •  Through Parameter flow variables •  Options list: glide_sif.py -help •  eg Core constraints: GlideLigandDocking-ref_ligand_file = <path/file.mae>

GlideLigandDocking-use_ref_ligand = yes

.../tmp/KNIME/knime_scripts_..._19009/glide_sif.py -forceplanar false -gridfile /tmp/recep.zip ... -poses_per_lig 1 ... -ref_ligand_file /tmp/ref.mae -use_ref_ligand yes

Prime Build Homology Modelling

•  Can use a specific template •  Via a parameter flow variable:

PrimeBuild--template-file pointing to the file or the column containing the template.

return “:Template:”;

•  KNIME in batch wrapper – list the allowed nodes and corresponding tags $SCHRODINGER/run -FROM maestro KNIME_batch.py <workflow.zip|workspace/workflow> -printallowed

Text Node Action

-------------------------------------------------------------------------------- Structure file Molecule Reader Read the first file specified in the node.

Selected entries Molecule Reader Read the entries that are selected (Maestro only) in the Project Table.

... Text file CSV Writer Write the output to specified file.

GUIsetting Double Input, Add a text box control to the panel Integer Input, and assign the value that is set in

String Input the text box.

•  -printusage, -printcmd, -run, -GUI options (already in 2013-1) See Shape screening workflow example: $SCHRODINGER/run KNIME_batch.py P\ Shape\ 1.zip -run -imae /tmp/fragments.mae -imae1 /tmp/Query.mae -omae /tmp/

PhaseShape.mae

KNIME in Batch Wrapper

Run PyMOL, Run Canvas

•  Run PyMOL –  Structures ungrouped to operate more easily on the structures –  Disabled when the workflow is executed in batch mode

•  Run Canvas –  Disabled when the workflow is executed in batch mode

Miscellaneous

•  Flow variables can be used in the Run Maestro node script (and already in Run Maestro command) See the Phase screening workflow example:

•  Parameter flow variables can be used for ConfGen •  Chemistry External Tool- cms as allowed input and output types Important bug fixes: •  Glide ligand docking – when multiple poses saved •  Extract mae properties – blank lists

2013 new features

•  Ease of use –  Stand-alone installation configuration in the Preferences

–  Improve the compatibility with other provider nodes (less conversions) –  Mine the list of the workflows and nodes in a workspace

–  Batch command generation – based on workflow annotations

•  Functionalities: –  Use backend option not exposed in the configuration panels –  SiteMap, PyMOL, Glide grid writer new nodes –  Easier prototyping with the new Chemistry External Tool node

•  Integration:

–  Run KNIME workflows from Seurat –  Can invoke web service to execute LigPrep node. Execute some Schrödinger backends through the Generic Web Service Client node

Some KNIME 2.6 and 2.7 new features

http://tech.knime.org/whats-new-in-knime-27 •  Random forest node •  Metanode – add/remove/reorder ports •  Grid layout and snapping •  Improved String Manipulation and Java Snippet nodes •  New database nodes: •  New quick form nodes:

Environment configurable in the preferences

•  For a stand-alone installation •  Other ways ($SCHRODINGER/run, start-up script, hunt based)

•  Set-up diagnosis node for the KNIME Server

–  Environment, scratch directories –  License, Backend installed

Batch command generation

“Structure file”

“Text file” “GUIsetting”

•  Command line to be run in batch generated based on node annotations Similar to the KNIME in Maestro ones eg “Structure file” for Molecule reader and writer nodes

eg $SCHRODINGER/run KNIME_batch.py QM.zip -printcmd $SCHRODINGER/knime -batch -nosave -maxThreads 1 -nosplash -workflowFile=QM.zip -option=8380,DataURL/0,/tmp/Aniline.smi,String -option=8386,output_file_name,/tmp/QMprotocolOutput.mae,String -option=8399,value/value,localhost:2,String

•  Workflows easily run in Seurat

KNIME in Seurat

Parameter flow variables

•  Any backend command line option not exposed in the node configuration panel •  Value, activate, value/structure from an input column, extra output column •  Metanode GUI with the Quick form nodes

eg SiteMap-sitebox = 3 SiteMap-ligmae = :CT lig:

Sitemap -HOST localhost:4 -j sitemap_-732349751_1 -maxsites 5 -modphobic 3 -keeplogs no -sitebox 3 -ligmae sitemap_-732349751_in_1_CT_lig.mae -prot sitemap_-732349751_in_1.mae

•  The console reads: Detected a SiteMap flow variable: SiteMap-sitebox=3 …

•  Glide ligand docking, Prime nodes, Protein preparation wizard, SiteMap, some Jaguar, MacroModel and Canvas nodes. Simple to activate for other nodes when needed.

Workflow examples and Workflow list node

•  Workflow examples available as a workspace under $SCHRODINGER/knime-v.../tutorial/ •  Workflow list node (free, in a separate plugin feature)

Lists the nodes and workflows in the workspace –  Latest modified workflows –  workflows containing specific nodes (eg example to create a new one) –  Compare several versions of a workflow (date, complexity) –  Find a workflow buried in groups –  Workspace clean up (size on disk)

SiteMap and Run PyMOL nodes

•  SiteMap –  Identify potential binding sites.

–  Evaluate a single binding site region (using the parameter flow variables)

•  Run PyMOL (free, in a separate plugin feature)

Standard input/ouput and Glide ligand docking node

•  More nodes input and output pdb and sdf. No need of converters. Glide ligand docking, Run Maestro command, Assign bond order, Split by Structure...

•  Extract automatically the generated properties Prime MM-GBSA and Glide ligand docking

•  Glide ligand docking with 1 output

SDF SDF SDF SDF SDF

New Chemistry external tool node

(Free, in a separate plugin feature) •  Optional input/output ports, output

column structure options, column name. •  Reads maegz files, input/output pdb,

output Surface type •  Flow variables, accessible by name •  Basename keyword, add extra columns to the output

Glide grid writer

•  Output file name + increment or from a column

•  See Docking > Grid generation 1-3 workflow example

Webservice

•  In the LigPrep node: ~/.schrodinger/schrodinger.hosts: name: pdx_web_server host: pdx-bld-l02.schrodinger.com port: 8080 processors: 3 •  Called from the Generic Webservice Client node For various tools with 1 input and 1 output eg LigPrep, QikProp

SCHRODINGER/ligprep -WAIT -HOST localhost:1 -nz -nc -s 4 -r 1 -epik -i 2 -W 'e,-ph,7.0,-pht,2.0'

-ismi ligprep_in.smi -osd ligprep_out.sdf

See General tools > Webservice workflow example

2013-1 new features

Based on KNIME 2.7 Infrastructure improvements •  Environment for a stand-alone installation configurable in the preferences.

Set-up diagnosis node.

•  Generation of the command line to be run in batch based on node annotations, batch execution setting panel. So workflows can easily be run in Seurat

•  Parameter flow variables to use any backend command line option not exposed in the node configuration panel

•  Workflow examples available in the installation

New nodes

•  SiteMap •  Run PyMOL

•  Workflow list – lists the nodes and workflows in the workspace

2013-1 New Features

New functionalities •  KNIME in Maestro – input structures from files

•  More nodes input and output pdb and sdf eg Glide ligand docking, Assign bond order, Split by Structure

•  Prime MM-GBSA and Glide ligand docking – Extract automatically the generated properties

•  Parameter flow variable – Glide ligand docking, Prime nodes, Protein preparation wizard, SiteMap, some Jaguar, MacroModel and Canvas nodes.

•  Glide ligand docking – with 1 output, sdf output type and other new functionalities •  Chemistry external tool – with optional input/output ports, access to the flow variables by name, basename

and other new keywords.

•  And many other fixes and minor improvements

2013-2 and 2013-3 New Features

2013-2 •  Latest version of KNIME (v2.8.2), included in 2013-3 •  Glide Grid Generation node- Save options •  Glide Multiple Grid Reader- Columns indicating original file location •  Run Maestro- Optionally provide command line arguments 2013-3 •  Easily specify custom configuration at application start •  Restored cross-platform 2D renderer functionality in the embedded KNIME installation •  Can invoke web service to execute LigPrep node •  Execute some Schrödinger backends through the Generic Web Service Client node •  New Glide grid writer node •  Flow variables can be used in the run Maestro command node script •  Expanded reporting from the Setup diagnosis node

2012 New Features – Infrastructure and Usability

Infrastructure •  Based on KNIME 2.5.1 •  Ported to Mac OS X •  Start-up script improvements including options to use a stand-alone KNIME

installation, specify the KNIME temporary directory. On Windows the bat window is silenced. Usability

•  The backend options reported in the console for all our nodes •  License checking mechanism improved to speed up the KNIME start-up •  Some nodes no long require a Schrödinger KNIME license: Reader, Writer, Converter, Scripting, KNIME-Maestro Connector, Run Maestro,

and Canvas


•  Node annotation - multiline, copy/paste, select all. Custom description via context menu •  Align nodes vertically •  Table view - select and copy a block of cells, sort data by clicking the column header (2 levels) New functionalities and nodes •  String Manipulation node - for operations on string columns •  JFreeChart nodes - new image output port (to be used in reports) •  Database nodes – new functionalities (schema browser, multi-line SELECT and non-SELECT

queries) Still not straighforward to update a database

KNIME 2.4- some New Features

•  Meta node - collapse, expand, rename •  Quick form node settings on a workflow or meta node level •  Auto-layout (experimental), fast node insertion •  Looping execution - step, pause, parallel execution. Column list loop start •  Shorter start-up time (nodes loaded on demand, batch execution)

New functionality and nodes: •  Free Marvin nodes •  XML processing nodes •  SDF Reader - multiple files, limit rows... CSVReader and Writer - handle gzipped files. Line

reader. •  GroupBy- new aggregation methods. Pivoting - more aggregation methods and multiple group

and pivoting columns. Colum to grid – group. •  Insert column header, Column Rename (Regex), Equal Size Sampling, Line Reader, Create

Temp Dir, Binner (Dictionary), Write PNG Image, Crosstab (contingency table)

Metanode with Quick Form Based Configuration Panel

•  The new Quick form nodes in KNIME 2.4 make possible to create a new “node” with a configuration dialog without requiring Java programming

•  This is actually a metanode exposing some configuration settings that can be passed to a Python script node as in the example below

Start-up Script - Options

Only affects the current session (don’t permanently alter the workspace settings in KNIME preferences)

•  -help .. Print information on command line options

•  -data <directory> ... Use <directory> as the KNIME workspace rather than the default of ~/workspace

•  -maxHeap <value> ... Set maximum heap size to <value> ex: -maxHeap 1024m

•  -maxThreads <value> ... Set maximum working threads to <value>

•  -defaultHost <value> ... Set default host from the schrodinger.hosts file (default is localhost)

•  -deleteTempFiles <value> ... Set option to delete temporary files (true or false)

•  -ooCmd <value> ... Set Open Office Spreadsheet command to <value>

•  -schrodingerTempDir <directory> ... Set Schrodinger temporary directory to <directory>.

If <directory> is "FromSchrodingerHost", then the value of 'tmpdir:' for 'localhost' in schrodinger.hosts file will be used.

•  -knimeTempDir <directory> ... Set KNIME temporary directory to <directory>.

If <directory> is "FromSchrodingerHost", then the value of 'tmpdir:' for 'localhost' in schrodinger.hosts file will be used.

•  -knimeInstallDir <directory> ... Set the KNIME installation directory to <directory>.

It can also be specified in the schrodinger.hosts file adding a line knime: <path> in the localhost section. This installation is then used in the Maestro workflows menu.

Batch execution specific options:

•  -batch ... Invoke the batch executor/engine

•  -nosave ... Do not save the workflow after execution has finished

•  -reset ... Reset workflow prior to execution

•  -preferences … path to the file containing eclipse/knime preferences

•  -workflowFile=<file> or -workflowDir=<dir> ... Workflow as zip or in a workspace

•  -destFile=<file.zip> or -destDir=<dir> ... Place where the executed workflow should be written (otherwise only saved in place)

•  -option=nodeID,name,value,type ... Set the option with name 'name' of the node with ID 'nodeID' to the given 'value', which has type 'type‘

•  And a couple others (see the –help message)

Start-up Script - Applications

•  Human readable option names (eg batch, memory) –  Instead of knime –Xmx4096m or in knime.ini

•  Define the Schrodinger environment and use the start-up script options with a stand alone installation (-knimeInstallDir), shell silenced on Windows

•  Use several KNIME versions, extensions in beta test (-knimeInstallDir) •  Easy file clean up (-knimeTempDir and -schrodingerTempDir) •  Use a workspace on different machines (-maxHeap -schrodingerTempDir -knimeTempDir -

maxThreads) and platforms (-ooCmd -defaultHost ) •  Make sure you didn’t forget to turn on Delete temporary files again (-deleteTempFiles true) •  Example: knime -maxHeap 4096m -max_threads 5 -defaultHost localhost -schrodingerTempDir /tmp/KNIME/ -knimeTempDir /tmp/ -deleteTempFiles true -ooCmd oocalc -data mainKNIMEworkspace/

2012 New Features – KNIME menu in Maestro

KNIME menu in Maestro under workflows Structure exchange

•  To send structures to and from an open KNIME session Run KNIME workflows

•  Build, import, edit and run workflows from Maestro on project table data. Input, output from Maestro specified by changing node names. Dynamically generated GUI to alter some parameters.

•  Rename and organize workflows in categories. They can also be installed under Scripts (running a workflow doesn’t required any knowledge about KNIME)

•  Disconnected calculations and progress tracking in Jobcontrol •  Workflow tools and examples available on the Workflow web page:

–  Process PDB codes, Split and align multimers, Binding site RMSD matrix –  Nearest neighbour similarity distribution, Substructure search –  others will be added to the workflow page (eg Compound set filtering)

KNIME – Maestro connection KNIMEusedinteractively

MaestroVisualinspection,selections... Server

ProjecttableStoretheresults

KNIME-Maestroconnector

Automatedtasks KNIME-Maestroconnector

Preprocessing

KNIME-Maestroconnector

Maestro - KNIME structure exchange

KNIME

Projecttable

Maestro Server http://www.schrodinger.com/knimeworkflows/Validatedockingparameters

Run workflows from Maestro



Splits and align multimers

Run from Maestro

2012 New Features – Functionalities

Functionalities •  Phase shape - ability to search a database •  New surface column type – created with the Python script node and displayed with Run Maestro

node •  Phase hypothesis and Glide grid types- display and access to the path •  Python script node new functions and access to the flow variables Many fixes and minor improvements •  Some nodes were renamed (eg Table Viewer, Run Spreadsheet Viewer)

Metanodes (available on the Workflow page) •  Run Maestro 1:1 •  Run PyMOL •  Jaguar pKa List of all the nodes •  Sitemap New workflow examples •  Glide grid writer

Some new metanodes and surface cell type

Some new metanodes and surface cell type

Splits and align multimers workflow example

Split a protein multimer by chain ID and align binding sites

•  Run on a set of multimers


Run on a set of structures (some of them may have only one chain)


Run from Maestro

2011 New Features

Infrastructure –  Windows installers –  Canvas 2D renderer on 64 bit KNIME workbench –  New start-up scripts option- memory limit, temporary directory… user/machine/OS-specific parameters

New nodes

–  Desmond system builder and Molecular dynamics –  Desmond trajectory manipulation and extract frames –  Desmond trajectory and CMS Readers –  Phase query on files (already possible on a Phase database) –  Canvas model building and prediction nodes (PLS and Bayes classification) –  Run Canvas –  Filter structures based on an ASL expression –  Calculate properties

2011 New Features

Node options and functionalities –  Prime Build homology model- include ligand and cofactors

–  Run Maestro- import groups, read Hypotheses, specify a project to add the structures to –  Protein preparation wizard, Epik, Molecular descriptors- job distribution

–  Ligprep- Implicit conversion

Even without a dedicated node

–  Search PDB header, Ligand Interaction Diagram (using the Chem. ext. tool node) –  Variable based readers (flow variables port added to each node)

–  Distance, angle or dihedral measurements (Python node)

Report designer

KNIME 2.2- some new features

New features •  Add flow variables port to each node. Enables control on execution order •  Allow for optional inputs. See Concatenate (Optional input) •  Loop-Concept for Chunking- implement streaming like approach

see also Schrodinger Row iterator loop start node (one row at a time) •  Support for workflow-local files (separate storage in workflow folder) •  Preconfigured Workflow example server

New Nodes •  Joiner (more flexible matching criteria, scalability, composite keys,...) •  Ungroup node- Split Collection in Rows •  Loop End node with two in/outputs •  CSV Reader - more flexible than File Reader node when input structure changes

(Schrodinger’s CSV reader node can read several files), Excel reader supports .xlsx •  In Labs: Web Analytics, Spotfire nodes, Modular Data Generation nodes •  Logistic Regression Learner & Applier, Item Set finder node ...

KNIME 2.3- some new features

•  Upgrade to Eclispe 3.6.1/BIRT 2.6/Java 1.6.0_22: enables KNIME to fully run on Mac OS X, no more xulrunner version problem

•  Sticky notes for workflow annotation •  Support for Hotkeys (e.g. to open a node's data output view) •  Data Flow Switches- IF/ENDIF, CASE/ENDCASE/ENDModelCASE nodes •  Preference option to omit the default node label ("Node 12") on node creation, to change the font

size for node name and label. Switch to turn off the node labels above the nodes (e.g. "File Reader") •  External tool- can be used as extension point (Java code to create simple interfaces) •  Ability to export KNIME views and KNIME workflows as SVG image •  Boolean DataType, Variable support for Database Reader and Database Connector and "Image" port to

carry, e.g. R views (needed for reporting)

KNIME 2.3- some new nodes and functionalities

•  Table creator- input the table content Create Table Structure- empty table with predefined structure

•  Java Edit variable, Java Expression based Row Filter and Row Splitter •  Column to Grid- split and turn a column to a grid to be used in a report table •  Read PNG Images, Image to Table and SVG Writer nodes •  List Files- list file locations •  Loop End (Column Append)- collecting columns instead of rows •  Extract Column Header into data row •  Excel Reader- speed improvements & user feedback on analyze •  A generic Cell replacer node •  Automatic Binner, Binner Apply nodes. Sparkline Appender node •  Java Snippet- Add Missing Value Handling, ability to throw exception (exits node execution) •  Variable support to be available in R nodes •  Extract System Properties (e.g. temp file, current user account)

Workflow annotations

Column to Grid, Read PNG images

Table creator

Node and workflow lists

Schrödinger Nodes

•  150+nodescoveringthewholeSchrodingerSuite

•  RunonLinux,MacandWindows

MolecularMechanics-MacroModelSinglePointEnergy,

Minimization,CoordinateScan-ConfGen-ConformationalSearch-Premin,Impref,UffminQuantumMechanics-JaguarSinglePointEnergy,

Minimization-NMRShieldingConstants-JaguarCharges

MolecularDynamics-DesmondSystembuilder-DesmondMolecularDynamics-Trajectoryextractframesand

manipulation-Trajectoryreader,CMSreaderCheminformaticsFingerprintBasedTools

-FingerprintGeneration-GeneratePairwiseMatrix-SimilarityMatrix-DissimilaritySelection-BuildReportandHierarchicalClusteringFiltersandMiningTools

-MCS-SubstructureSearch-Structure,REOSFiltersUtilitiesandConverters

-PCA,MDS-  CombineFingerprints,Concatenate

Bitvectors,ConvertFingerprinttoBitvector,etc

Modeling-BayesModelBuilding,prediction-PLSModelBuilding,PredictionPharmacophoreModeling-PhaseShape-PhaseDBQuery,FileQuery-PhaseDBCreation-PhaseHypothesisIdentificationCombinatorialLibraries-CombiGlideReagentPreparationand

LibraryEnumeration-FragmentsfromMoleculesandjoinerDockingandScoring-GlideGridGeneration-GlideLigandDocking,Ensemble

Docking-  XPVisualizer

Post-processing-PrimeMM-GBSA-EmbraceMinimization-StrainRescore,PoseEntropy-PoseFilter,GlideMerge,SortResults-GlideEnsembleMergeProteinStructurePrediction-BLAST-PrimeEnergy-PrimeBuildHomologyModel-PrimeSideChainSampling,

Minimization-IFDandindividualsteps

Schrödinger Nodes

ProteinPreparation-ProteinPreparationWizard-ProteinAssignmentLigandPreparation-LigPrepandtheindividualsteps-EpikPropertyGeneration-QikProp,MolecularDescriptors-CanvasdescriptorsFiltering-Ligfilter,Ligparse,PropfilterScripting-RunMaestroCommand-ChemistryExternaltoolnode-PythonScriptnodes

Reporting-RunMaestro,RunPyMOL,RunCanvas-MultipleSequenceViewer-KNIME-Maestroconnector-Tableviewer,SpreadsheetviewerTools-SiteMap-Residuescanning/affinitymaturation-Conformercluster-Workflowlist-SetupdiagnosisDataManipulation-CompareLigands-LookupandAddColumns-GroupandUngroupMAE

StructureManipulation-AddHydrogens-DeleteAtoms-SplitbyStructure

-MAEatom/bondpropertyParser-Extract,SetandDeleteMAEProperties-SetMoleculeTitle,MAEIndexUtilities-GetPDB-AlignBindingSites-ProteinStructureAlignment-Rotateall-RMSD-AssignBondOrders-UniqueTitleCheck-CheckPDBName-SDFormatChecker-GenerateSmarts,UniqueSmiles-RRHOEntropy-BoltzmannPopulation-VolumeOverlapMatrixReaders/Writers-CSVReader(severalinputs)-MoleculeReaderandWriter

-SD,PDB,Mol2Readernodes-Sequence,AlignmentReadersand

Writers-FingerprintReaderandWriter-HypothesisReaderandWriter-GlideGridReaderandWriter-VariableBasedGlideGridReaderConverters-String-to-Type-Molecule-to-MAE,MAE-to-Pdb,to-SD,

to-Smilesandto-Mol2,SD-to-Smiles-PoseViewer-to-Complexesand

Complexes-to-PoseViewers-Hartree-to-kcal/molandkJ-to-kcal

Parameter Flow Variable

The Parameter flow variable functionality is supported for the following nodes:

•  LigPrep •  Glide Ligand Docking

•  Glide Grid Generation •  Jaguar Minimization

•  Jaguar NMR Sheilding

•  Jaguar Single Point Energy •  Shape Screening

•  Protein Preparation Wizard •  SiteMap

•  ConfGen

•  MacroModel Conformational Search •  MacroModel Minimization

•  MacroModel Single Point Energy •  Prime Build Homology

•  Prime Energy •  Prime MMGBSA

•  Prime Minimization

•  Bioluminate Residue scanning/Affinity maturation •  Substructure Search

•  Canvas Fingerprint Generation •  Split Structure

•  Conformer Cluster

•  Molecule reader •  Molecule to Mae

KNIME Workflow Page http://www.schrodinger.com/knimeworkflows/

Cheminformatics •  Substructure Search •  Clustering, diversity selection, similarity

search •  Database analysis, MCS Docking and post-processing •  Binding site (SiteMap, grid generation) •  Docking, Virtual screening, Ensemble

docking •  Loop over docking parameters •  Validate docking parameters (using the

Maestro connector node) Pharmacophore modeling •  Phase Shape screening , hypothesis

identification and database screening Molecular Mechanics •  Compare conformational search

methods •  Conformational search

Quantum mechanics •  ESP charges, Jaguar pKa, QM

properties, Semiempirical •  Conformational search and QM

refinement (using the Report designer) Molecular Dynamics: Desmond simulation Library design: Library enumeration Protein design •  Model building and refinement •  Induced Fit Docking •  Bioluminate •  Split and align multimers Real World Examples •  Vendor database preparation •  SiteMap and docking •  Binding site shape clustering •  Protein model building •  Protein structural diversity analysis •  Co-crystallized ligand docking

Labs: Parameter flow variables General tools •  Python script, Chemistry external tool,

Run maestro command node use-cases…

•  Run PyMOL •  Workflows in the current workspace •  KNIME-Maestro connector •  Installation update/creation •  Batch execution KNIME desktop: GroupBy, loop examples

… Simple examples: Epik, conformer cluster,

Multiple Sequence Viewer, Run (command) Maestro, Free nodes, Readers and writers, Commonly used workbench nodes

Most exciting, simplest to start with

Binding shape site clustering and ensemble docking

Osguthorpe,D.J.;Sherman,W;Hagler,A.T.ChemBiolDrugDes2012;80:182–193

Cocrystallized Ligand Redocking

•  This workflow is derived from the work published by Paulette Greenidge from Novartis in: •  Improving Docking Results via Reranking of Ensembles of Ligand Poses in Multiple X-ray Protein

Conformations with MM-GBSA. Greenidge, P. A.; Kramer, C.; Mozziconacci, J.C.; Sherman, W.; J. Chem. Inf. Model. 2014, 54(10), 2697–717

•  MM/GBSA binding energy prediction on the PDBbind data set: Successes, failures, and directions for further improvement. Greenidge, P. A.; Kramer, C.; Mozziconacci, J.C.; Wolf, R.M.; J. Chem. Inf. Model. 2013, 53(1), 201–9

Schrödinger KNIME Extensions

KNIMEExtensionsProductManager: Jean-ChristopheMozziconacci

([email protected])

VicePresidentofTechnology: VolkerEyrich

MainDeveloper: RavikiranKuppuraj

andthePyDevdevelopmentteam

QA: SimonFoucher

Workflowexamples: TanviBhola

TechnicalSupport: KatalinPhimister,PavelGolubkov

Marketing: JarredYacob

Documents

Schrödinger KNIME Extensions